DCE-MRI based voxelized computational model for chemotherapeutic drug transport in human brain tumor

Abstract

Understanding the distribution of chemotherapeutic drugs in tumor when administered systemically is important for assessing the efficiency of anticancer treatment. Despite numerous researches in this field, prediction of drug distribution within the tumors still remains a challenge due to wide heterogeneity of tumor vasculature. In this work, a voxelized computational fluid dynamics (CFD) model based on dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) has been employed to simulate the transport of chemotherapeutic drug (doxorubicin) in human brain tumors. The DCE-MRI data provide realistic heterogeneous vasculature of a tumor and help in determining the permeability of the tissue vascular wall to contrast agent and interstitial volume fraction (porosity) of the tissue. The permeability of the tissue to doxorubicin is calculated by correlating it with the permeability of the tissue to contrast agent. Patient-specific arterial input function (AIF) estimation has also been done to make the model patient-specific. The computational model provides steady-state interstitial fluid pressure and velocity values, which are further used to predict transient drug concentration in tumor tissue. Simulation results show that doxorubicin accumulates more in high permeability areas initially. However, drug accumulation increases in areas of higher interstitial volume fraction as time progresses. This observation implies that higher accumulation of chemotherapeutic drugs occurs in higher porosity areas, facilitating larger tumor cell killing in those areas.